Power factor correction is one of the most routine investments in industrial electrical engineering — install a capacitor bank, reduce reactive power charges, improve the utility bill. For decades this approach worked reliably. In a world dominated by linear inductive loads, capacitors did exactly what they were supposed to do.
That world no longer exists in most industrial facilities. Variable speed drives, switch-mode power supplies, uninterruptible power supplies, and other non-linear loads now represent a significant and growing proportion of the electrical demand in manufacturing, water treatment, commercial HVAC, and virtually every other sector. In these environments, the straightforward relationship between capacitors and power factor breaks down — and the consequences can be severe.
The core problem is parallel resonance. When a capacitor bank is connected to a distribution system containing transformer inductance, the two elements form a resonant circuit whose natural frequency depends on the capacitor size and the network short-circuit level. If this resonant frequency coincides with a characteristic harmonic of the non-linear loads — the 5th, 7th, or 11th harmonic of a 6-pulse drive installation — the results are dramatic. A small harmonic current injected by a variable speed drive excites the LC circuit into oscillation. Inside the resonant loop, between the capacitor and the transformer, circulating currents build to 20 to 40 times the injected harmonic current. Fuses blow. Capacitor cases bulge. Transformers run hot. Protection relays trip for no apparent reason. Replace the fuse and it blows again — because the resonant condition that caused it has not been addressed.
What makes this problem particularly insidious is that it is largely invisible to conventional instrumentation. The currents circulating inside the LC loop do not appear at the supply meter. An engineer reviewing power quality data sees nothing obviously wrong. And the resonant condition can develop gradually as drives are added to an existing installation, or abruptly when a capacitor bank is expanded — turning a previously safe system into a dangerous one with a single switching operation.
The new article in the IPQDF Technical Reference Series provides a complete engineering treatment of this subject. It covers the physics of parallel resonance and why the traditional current divider intuition breaks down at resonance, the field symptoms that identify a resonance problem, a six-step assessment methodology for safe capacitor specification, detuned capacitor banks and what they do and do not achieve, passive harmonic filter design and its limitations, and active harmonic filters and how they eliminate the resonance risk entirely. A practical selection guide with decision flowchart, technology comparison table, and worked example help the practitioner choose the right solution for their specific installation.
The article includes three interactive figures — including a resonance explorer with real-time impedance and capacitor current calculations based on actual 6-pulse VFD harmonic injection levels — that make the physics tangible and directly applicable to real installations.
If you have ever had a capacitor bank fail in a plant with variable speed drives, or if you are about to specify power factor correction for such an installation, this article is the reference you need.
Denis Ruest · IPQDF Technical Reference Series · Harmonics and Power Factor Capacitors: Understanding Failure, Resonance and the Filter Solution
Content drafted with AI assistance and validated by the author based on 30 years of experience in the Power Quality field.
© 2026 Denis Ruest — International Power Quality Discussion Forum (IPQDF). Reproduction permitted for non-commercial educational purposes with full attribution to the author and a link to the original article at ipqdf.com.